The development and application of sensitive methods for the determination of hydrogen peroxide led, a few years ago in the laboratories of the Johnson Research Foundation, to the recognition, of intact mitochondria as an effective source of H₂O₂ (Fig. 1; refs. 1–4). Previous observations by Jensen (5) and by Hinkle et al. (6) had indicated that the mitochondrial respiratory chain was capable of producing H₂O₂. However, these results were taken with caution, in the sense that they might reflect an artificial activity induced by the ultrasonic or alkaline treatment used in the preparation of the submitochondrial particles. In 1971, Chance and Oshino (1) demonstrated variations in the level of the catalase intermediate of the peroxisomal-mitochondrial fraction of rat liver following the addition of mitochondrial substrates and uncouplers. In the same year, Loschen et al. (2) showed H₂O₂ formation in pigeon heart mitochondria and its relationship to the mitochondrial metabolic state by using the peroxidase-scopoletin method. It was realized that this assay could be easily interfered by endogenous hydrogen donor of the horseradish peroxidase and by exogenous hydrogen donors in the mitochondrial preparations, and consequently, an alternative method was developed.